projek_padi/web_TA/MODEL_API_SETUP.md

📊 Laravel Model & Database Setup - Classification System

✅ Apa yang Sudah Dibuat

1. Model (Classification.php)

• Represents classifications table
• Fillable properties untuk menyimpan hasil klasifikasi
• JSON casting untuk all_predictions field

2. Migration (create_classifications_table.php)

• Table schema dengan columns:
  • id - Primary key
  • image_path - Path gambar yang disimpan
  • filename - Original filename
  • predicted_class - Hasil prediksi (Bacterialblight, Brownspot, Leafsmut)
  • confidence - Score confidence (0-1)
  • all_predictions - JSON semua prediksi
  • disease_name - Nama penyakit dari disease info
  • severity - Tingkat severity penyakit
  • notes - Catatan tambahan
  • created_at, updated_at - Timestamps

3. Controller (ClassificationHistoryController.php)

• Index: Get semua classification dengan pagination
• Show: Get detail classification tertentu
• Destroy: Delete classification + image
• Stats: Get statistik klasifikasi

4. Updated Files

• ClassificationController.php - Save data ke model
• routes/api.php - Add history endpoints
• database/migrations/2026_03_08_182722_create_classifications_table.php - Table schema

---

🔌 API Endpoints

Classification Endpoints

1. Test Connection

GET /api/classification/test

Response:

{
  "success": true,
  "message": "Koneksi ke model API berhasil",
  "model_info": { ... }
}

2. Classify Image (Analyze Only)

POST /api/classification/classify
Content-Type: multipart/form-data

Body:
  image: <file>

Response:

{
  "success": true,
  "data": {
    "predicted_class": "Bacterialblight",
    "confidence": "95.23%",
    "disease_info": {...},
    "timestamp": "2026-03-09T..."
  }
}

3. Classify & Save Image

POST /api/classification/classify-and-save
Content-Type: multipart/form-data

Body:
  image: <file>
  notes: "Optional notes"

Response: Same as above + saves to database

---

History Endpoints

4. Get All Classifications (Paginated)

GET /api/classifications?page=1

Response:

{
  "success": true,
  "data": {
    "data": [
      {
        "id": 1,
        "image_path": "classifications/...",
        "filename": "image.jpg",
        "predicted_class": "Bacterialblight",
        "confidence": 0.95,
        "disease_name": "Bercak Bakteri",
        "severity": "Sedang hingga Tinggi",
        "notes": "...",
        "created_at": "2026-03-09T10:30:00Z"
      }
    ],
    "current_page": 1,
    "total": 50,
    "per_page": 15
  }
}

5. Get Classification Detail

GET /api/classifications/{id}

Response:

{
  "success": true,
  "data": {
    "id": 1,
    "image_path": "...",
    "filename": "...",
    "predicted_class": "...",
    "confidence": 0.95,
    "all_predictions": {
      "Bacterialblight": 0.95,
      "Brownspot": 0.04,
      "Leafsmut": 0.01
    },
    "disease_name": "...",
    "severity": "...",
    "notes": "...",
    "created_at": "...",
    "updated_at": "..."
  }
}

6. Delete Classification

DELETE /api/classifications/{id}

Response:

{
  "success": true,
  "message": "Classification deleted successfully"
}

7. Get Statistics

GET /api/classifications/stats/summary

Response:

{
  "success": true,
  "data": {
    "total_classifications": 50,
    "average_confidence": "93.45%",
    "by_disease": [
      {
        "disease_name": "Bacterialblight",
        "count": 30
      },
      {
        "disease_name": "Brownspot",
        "count": 15
      },
      {
        "disease_name": "Leafsmut",
        "count": 5
      }
    ]
  }
}

---

📦 Database Schema

CREATE TABLE classifications (
  id BIGINT PRIMARY KEY AUTO_INCREMENT,
  image_path VARCHAR(255) NULLABLE,
  filename VARCHAR(255) NOT NULL,
  predicted_class VARCHAR(255) NOT NULL,
  confidence FLOAT NOT NULL,
  all_predictions JSON NOT NULL,
  disease_name VARCHAR(255) NULLABLE,
  severity VARCHAR(255) NULLABLE,
  notes TEXT NULLABLE,
  created_at TIMESTAMP,
  updated_at TIMESTAMP
);

---

📁 File Structure

app/
├── Http/
│   └── Controllers/
│       ├── ClassificationController.php       (✏️ Updated)
│       └── ClassificationHistoryController.php (✨ NEW)
└── Models/
    └── Classification.php                     (✨ NEW)

database/
└── migrations/
    └── 2026_03_08_182722_create_classifications_table.php (✨ NEW)

routes/
└── api.php                                    (✏️ Updated)

---

🚀 Usage Examples

Save Classification from Controller

use App\Models\Classification;

// Automatic save in ClassificationController
Classification::create([
    'image_path' => 'classifications/image.jpg',
    'filename' => 'rice_leaf.jpg',
    'predicted_class' => 'Bacterialblight',
    'confidence' => 0.95,
    'all_predictions' => ['Bacterialblight' => 0.95, ...],
    'disease_name' => 'Bercak Bakteri',
    'severity' => 'Sedang hingga Tinggi',
    'notes' => 'Classification from mobile app',
]);

Query Classifications

// Get latest 10 classifications
$classifications = Classification::orderBy('created_at', 'desc')
    ->limit(10)
    ->get();

// Get by disease
$bacterialblight = Classification::where('disease_name', 'Bercak Bakteri')->get();

// Get high confidence classifications
$highConfidence = Classification::where('confidence', '>=', 0.9)->get();

Delete Classification with Image

$classification = Classification::find(1);

// Delete image
if ($classification->image_path) {
    Storage::disk('public')->delete($classification->image_path);
}

// Delete record
$classification->delete();

---

✅ Migration Status

$ php artisan migrate

   INFO  Running migrations.  
  2026_03_08_182722_create_classifications_table ............... 347.26ms DONE

✅ Table classifications berhasil dibuat!

---

🔍 Verification

Check Table Structure

php artisan tinker

>>> DB::table('classifications')->getColumns()
// Returns table structure

>>> App\Models\Classification::count()
// Returns number of classifications

Check Fillable Properties

php artisan tinker

>>> (new App\Models\Classification())->getFillable()
// Returns: ['image_path', 'filename', 'predicted_class', ...]

---

📊 Data Flow

Mobile App
    ↓
POST /classify-and-save (with image)
    ↓
ClassificationController
    ↓ Call Python API
    ↓ Get result
    ↓
Classification::create([...])  ← Save to DB
    ↓
Return JSON response
    ↓
Mobile App displays result

---

🎯 Next Steps

1. ✅ Model & Migration created
2. ✅ Database table created (php artisan migrate)
3. ✅ History endpoints added
4. → Test API endpoints dengan Postman/curl
5. → Update mobile app untuk handle response
6. → Create dashboard untuk view classifications

---

🧪 Test API Endpoints

Test Save Classification

curl -X POST http://127.0.0.1:8000/api/classification/classify-and-save \
  -F "image=@rice_leaf.jpg" \
  -F "notes=Test classification"

Test Get History

curl http://127.0.0.1:8000/api/classifications

Test Get Stats

curl http://127.0.0.1:8000/api/classifications/stats/summary

---

📝 Notes

• ✅ All classifications auto-save to database
• ✅ Images stored in storage/app/public/classifications/
• ✅ JSON data properly casted in model
• ✅ Pagination on history endpoint (15 per page)
• ✅ Delete endpoint includes image cleanup
• ✅ Statistics endpoint aggregates data

---

Status: ✅ COMPLETE
Tables Created: 1 (classifications)
Controllers: 2 (ClassificationController, ClassificationHistoryController)
API Endpoints: 7 total
Last Updated: March 9, 2026